Slashdot Mirror

It sounds like you really don't know about what they call the "security chip". That's the secret ingredient. You can bypass it if you can do a little hacking, but that is the "fairy dust" which prevents "completely standard Intel-based hardware" from running Mac OS. (And therefore it is specifically *not* completely standard.)

"Easily run Windows 7" is not a feature, it is a bug. Windows is as good of an operating system as a pile of donkey shit on the bottom of the ocean. If you want to run Windows, you don't need Apple hardware, you simply need to kill yourself. And if you want to run Linux, then you are a rockstar who doesn't need to pay for Apple hardware to have a good computer experience. For the rest of people, they pay the cost to run the best consumer-oriented OS.

Not sure how different the port is from one proc to the other, but someone has the x86 version of OPENSTEP emulated on Mac OS X in Parallels. BasiliskII was the gold standard of 68k emus, but I haven't heard of anyone running anything but MacOS on it (I had researched a little trying to find a way to emulate A/UX, but the math coprocessor isn't emulated, so no dice).

EVERYTIME, which is way to frequent, that my mac book Pro crashes, running the monolithic microkernel MacOSX (10.4 - 10.5.1), the Mac asks if I want to report it to Apple. I always say yes and berate them for having IO Drivers in the Kernel (since that's been the reason for the crashes - every time. I encourage them to remove the drivers from the their monolithic microkernel and put them into their own protected space.

Someone has to tell Apple that their Kernel design implementation sucks big time - simply because it crashes WAY TOO OFTEN. MacOSX 10.4 through the current up to date Mac OSX 10.5.1 Leopard (System Version: Mac OS X 10.5.1 (9B18), Kernel Version: Darwin 9.1.0) crashes very frequently week with IO problems (see the actual crash logs at the end of this posting). Usually these problems are with device drivers such as EyeTV or Parallels. If this was a true microkernel design like Minix or QNX the entire machine would not have to be rebooted, just the EyeTV or Parallels apps would need to be rebooted. Why the heck should a problem with the USB driver bring down the entire OS? How on earth can that be justified Apple? It can't so improve the quality by removing ALL drivers from the Kernel and put them in their own processes. Thank you very much in advance.

APPLE PLEASE PLEASE PLEASE REMOVE ALL DRIVERS FROM THE MACH XNU MICROKERNEL AS SOON AS POSSIBLE. THANKS. AT LEAST GIVE ME THE CHOICE OF HAVING THEM SEPARATED - THE EXTRA CPU % COST IS A PRICE THAT I AS A USER WOULD MAKE TO GAIN THE FAULT TOLERANCE AND RELIABILITY. Below you'll see my actual kernel crash logs for six months - as you can see the number of crashes are intolerable and just amazing to behold when a true microkernel that separates out drivers would have prevented reboots in all these cases.

I HEREBY Challenage ALL Mac OSX users to publish their Kernel Crash Logs for ALL the world to see. Maybe this way APPLE will take microkernels seriously.

Links to the issue of the flawed XNU kernel. (Gee, XNU, almost sounds like the XeNU character out of the Scientology creation mythology. XeNU http://en.wikipedia.org/wiki/Xenu. ;--).

The culprit: bad monolithic design of http://en.wikipedia.org/wiki/XNU#I.2FO_Kit and http://en.wikipedia.org/wiki/Mach_kernel

http://www.linuxjournal.com/article/6105
http://www.maconintel.com/news.php?article=177
http://developer.apple.com/macosx/architecture/index.html
http://www.applematters.com/index.php/section/comments/how-long-will-apple-keep-the-mach-microkernel/

"Frankly, I think it's a piece of crap," Torvalds says of Mach [XeNU], the microkernel on which Apple's new operating system is based. "It contains all the design mistakes you can make, and manages to even make up a few of its own." - Linus Torvalds, http://news.zdnet.co.uk/software/0,1000000121,2085525,00.htm

I only quote Linus because he's right regarding MACh XeNU. However, he's wrong about microkernels in general as the frequent crashing of Linux reveals.

---- ACTUAL MacBook Pro Monolithic XNU Kernel Crash Logs ----- REAL WORLD CRASHES REVEALED -----

Sat Mar 24 07:38:10 2007
panic(cpu 0 caller 0x0035AE53): freeing free mbuf
Backtrace, Format - Frame : Return Address (4 potential args on stack)
0x36563ca8 : 0x128d08 (0x3c9ac4 0x36563ccc 0x131de5 0x0)
0x36563ce8 : 0x35ae53 (0x3ea228 0x9cfc 0x36563d28 0x2)
0x36563d08 : 0x35b1f3 (0x4835b800 0x804c 0x36563d28 0x800)
0x36563d28 : 0xa3d6ad (0x4835b800 0x36563dec 0x6 0x6007c0e0)

Hell, I despise Apple.

Hate it. Hate it. HATE IT. Jobs and his RDF can suck my hairy white...nevermind...

But even I understand the reason for BootCamp.

If Apple didn't, someone else would.

Apple is legendary for their death-grip on platform experience.

People were coming up with ugly hacks just to get Windows on their Mac systems. Some resulting in brickified systems (which the user is going to at least TRY to return.

In short, people had hardware that WOULD run Windows. So they're gonna try. Even if it's just to say "I can run both! NYAH!"

So what makes more sense to Apple? To keep processing returns for people who kill their systems fucking around in EFI? Or to release a safe(er) product that'll allow people to get their XP on without murdering their system?

As to why Apple didn't do it in the first place? They wanted to see what the demand was in the early adopters. If nobody tried to put Windows on their systems, Apple doesn't waste time with BootCamp. If they do, Apple releases BootCamp and (because of the frothing fanboyism in the RDF) they're hailed as offering the users a choice (even though they didn't do so from the get-go).

http://www.maconintel.com/news.php?article=29 ( MacOnIntel link )

blakespot

DLightman

blakespot

blakespot